Contents Menu Expand Light mode Dark mode Auto light/dark mode
Spinal Cord Toolbox documentation
Logo

Overview

  • Introduction
  • SCT Concepts
    • PAM50 Template
    • Contrast-specific vs. contrast-agnostic
    • Point labeling conventions
    • Inspecting the results of your analysis (Quality Control, FSLeyes)
    • Voxels Space Orientation and Coordinate Conventions
    • Temporary Directories
    • Warping fields
  • Testimonials
  • Studies using SCT

User section

  • Installation
    • Installation for MacOS
    • Installation for Linux
    • Installation for Windows
  • Getting Started
  • SCT Courses
  • Tutorials
    • Segmentation
      • Before starting this tutorial
      • Contrast-specific vs. contrast-agnostic
      • Hands-on: Using sct_deepseg on T2 data
      • sct_deepseg: Other specialized models
    • Vertebral labeling
      • Before starting this tutorial
      • Types of vertebral labels
      • Point labeling conventions
      • Labeling algorithm: sct_label_vertebrae
      • Applying the labeling algorithm
      • Alternative #1: Manually labeling the C2-C3 disc
      • Alternative #2: Manual labeling all labels
      • How many vertebral labels should I use for registration?
      • Extracting specific labels for registration
    • Shape analysis
      • Compute CSA (and other shape metrics)
        • Before starting this tutorial
        • Cross-sectional area (CSA)
        • Other shape metrics
        • Verify the correctness of the metrics
      • Quantify spinal cord compression
        • Before starting this tutorial
        • Normalization pipeline
        • Generate the necessary input files
        • Compute normalized morphometrics
    • Lesion analysis
      • Before starting this tutorial
      • Lesion segmentation in spinal cord injury (SCI)
      • Lesion segmentation in multiple sclerosis (MS)
      • Compute lesion morphometric measures
      • Template/Atlas-based lesion analysis
    • Spinal nerve rootlets segmentation
      • Before starting this tutorial
      • Spinal nerve rootlets segmentation
    • Registration to template
      • Registering labeled anatomical images to the PAM50 template
        • Before starting this tutorial
        • Registration algorithm: sct_register_to_template
        • Applying the registration algorithm
        • Customizing the registration command
        • Transforming the template using warping fields
      • Coregistering additional data (MT, DT) to the PAM50 template
        • Before starting this tutorial
        • Spinal cord segmentation for MT1 data
        • Creating a mask around the segmentation
        • Registration Option 1: Reusing previous warping fields
        • Registration Option 2: Direct registration to the template
        • Transforming the template using warping fields
      • Registering lumbar images to the PAM50 template
        • Before starting this tutorial
        • Using sct_deepseg to segment the lumber region of the spinal cord
        • Adding landmark labels for template matching
        • Applying the registration algorithm
    • Multimodal registration
      • Computing MTR using co-registration between MT0 and MT1 data
        • Before starting this tutorial
        • Spinal cord segmentation for MT1 data
        • Creating a mask around the segmentation
        • Coregistering MT0 with MT1
        • Computing MTR using coregistered MT data
      • Contrast-agnostic registration with deep learning
        • Before starting this tutorial
        • Preprocessing steps to highlight the spinal cord (T2w)
        • Preprocessing steps to highlight the spinal cord (T1w)
        • Coregistering T1w with T2w
    • Gray matter segmentation
      • Segmenting the gray and white matter for T2* data
        • Before starting this tutorial
        • Gray matter segmentation algorithm: sct_deepseg_gm
        • Applying the gray matter segmentation algorithm
        • Computing the white matter segmentation
      • Computing metrics using GM and WM segmentations
        • Before starting this tutorial
        • Using binary masks to compute CSA for gray and white matter
        • Using binary masks to extract intensity values for gray and white matter
      • Improving registration results using white and gray matter segmentations
        • Before starting this tutorial
        • GM-informed registration between the PAM50 template and T2* data
        • Reusing the GM-informed warping field to improve MTI registration
    • Atlas-based analysis
      • Before starting this tutorial
      • Atlas-based analysis
      • The partial volume effect
      • Overcoming the partial volume effect
      • Transforming the GM/WM atlas to the MT space using warping fields
      • Using the atlas to extract MTR in white matter
      • Using the atlas to extract MTR from specific white matter tracts
      • Modifying info_label.txt to add custom tracts to your analysis
    • Diffusion-weighted MRI
      • Before starting this tutorial
      • Preprocessing steps to highlight the spinal cord
      • Motion correction for dMRI images
      • Registering dMRI data to the PAM50 template
      • Computing DTI for motion corrected dMRI data
      • Extracting DTI from specific spinal cord regions
    • Functional MRI
      • Before starting this tutorial
      • Preprocessing steps to highlight the spinal cord
      • Motion correction for fMRI images
      • Registering fMRI data to the PAM50 template
      • Spinal labeling
      • Warping the spinal levels to the fMRI space
    • Other features
      • Spinal cord smoothing as a preprocessing operation
      • Visualizing misaligned cords with 2D sagittal flattening
    • Analysis pipelines with SCT
      • Before starting this tutorial
      • Introduction to building processing pipelines using scripts
      • Running the sample script (process_data.sh) using sct_run_batch
      • Inspecting the results of processing
      • What if things go wrong?
  • Command-Line Tools
    • Segmentation
      • sct_create_mask
      • sct_deepseg
        • spinalcord
        • sc_t2star
        • seg_ms_sc_mp2rage
        • sc_epi
        • sc_mouse_t1
        • sc_lumbar_t2
        • graymatter
        • gm_wm_exvivo_t2
        • gm_sc_7t_t2star
        • gm_mouse_t1
        • gm_wm_mouse_t1
        • lesion_sci_t2
        • seg_sc_ms_lesion_stir_psir
        • lesion_ms_axial_t2
        • lesion_ms_mp2rage
        • lesion_ms
        • tumor_edema_cavity_t1_t2
        • tumor_t2
        • rootlets
        • sc_canal_t2
        • totalspineseg
      • sct_deepseg_gm
      • sct_deepseg_lesion
      • sct_deepseg_sc
      • sct_get_centerline
      • sct_propseg
    • Segmentation analysis
      • sct_analyze_lesion
      • sct_compute_hausdorff_distance
      • sct_compute_compression
      • sct_detect_compression
      • sct_dice_coefficient
      • sct_process_segmentation
    • Labeling
      • sct_detect_pmj
      • sct_label_vertebrae
      • sct_label_utils
    • Registration
      • sct_apply_transfo
      • sct_get_centerline
      • sct_register_multimodal
      • sct_register_to_template
      • sct_straighten_spinalcord
      • sct_warp_template
    • Diffusion MRI
      • sct_dmri_compute_bvalue
      • sct_dmri_concat_bvals
      • sct_dmri_concat_bvecs
      • sct_dmri_compute_dti
      • sct_dmri_denoise_patch2self
      • sct_dmri_display_bvecs
      • sct_dmri_moco
      • sct_dmri_separate_b0_and_dwi
      • sct_dmri_transpose_bvecs
    • Magnetization transfer
      • sct_compute_mtr
      • sct_compute_mtsat
    • Functional MRI
      • sct_fmri_compute_tsnr
      • sct_fmri_moco
    • Metric processing
      • sct_analyze_texture
      • sct_extract_metric
    • Image manipulation
      • sct_convert
      • sct_crop_image
      • sct_denoising_onlm
      • sct_flatten_sagittal
      • sct_image
      • sct_maths
      • sct_merge_images
      • sct_resample
      • sct_smooth_spinalcord
    • Miscellaneous
      • sct_compute_ernst_angle
      • sct_compute_snr
      • sct_download_data
      • sct_qc
      • sct_run_batch
    • System
      • sct_check_dependencies
      • sct_version
    • Help text (-h) for all tools
  • Analysis pipelines
  • FSLeyes Integration
  • Help
  • Citing SCT

Developer section

  • Contributing to SCT
  • Changelog
  • License
Back to top
Edit this page

Contrast-agnostic registration with deep learningΒΆ

This tutorial will demonstrate how to coregister two images together that have different contrasts using deep learning. The algorithm is based on SynthMorph. More details of its implementation in SCT can be found here.

  • Before starting this tutorial
  • Preprocessing steps to highlight the spinal cord (T2w)
  • Preprocessing steps to highlight the spinal cord (T1w)
  • Coregistering T1w with T2w
Next
Before starting this tutorial
Previous
Computing MTR using coregistered MT data
Copyright © 2020, SCT Contributors
Made with Sphinx and @pradyunsg's Furo